From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ursula Braun Subject: Re: [PATCH V6 net-next 2/2] smc: introduce socket family AF_SMC Date: Fri, 25 Sep 2015 09:27:13 +0200 Message-ID: <1443166033.6021.6.camel@BR9GV9YG.de.ibm.com> References: <1442227366-79236-1-git-send-email-ubraun@linux.vnet.ibm.com> <1442227366-79236-3-git-send-email-ubraun@linux.vnet.ibm.com> <20150921.114425.119948328742744162.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20150921.114425.119948328742744162.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-Archive: List-Post: To: David Miller Cc: utz.bacher@de.ibm.com, netdev@vger.kernel.org, linux-s390@vger.kernel.org, schwidefsky@de.ibm.com, heiko.carstens@de.ibm.com, ursula.braun@de.ibm.com List-ID: On Mon, 2015-09-21 at 11:44 -0700, David Miller wrote: ... > This is a huge and complex submission and I'm already burnt out > reading the changes thus far. > > You have a lot to fix up and you can expect many revisions to be > necessary before these changes are ready for integration upstream. > And that's if you are lucky and someone actually continues to review > this work. > Hi Dave, I appreciate the time you have already spent analyzing my SMC code. It is large and complex, and I understand that you are not willing to spend the time to review the overall code in future iterations. Before me spending more time on SMC, I need a hint about *your* preferred way to submit that large piece of code in general, and about the overall future of SMC: - Supposed the SMC code is improved to an acceptable quality and maybe even gets additional reviewers, are you willing to accept the code at all, given it is self-contained (our own can of worms, as you said)? - I realized that I have to split up the large chunk of code into smaller patches. Do you prefer going with a first minimal self contained patch set first, providing basic communication capabilities, and then incrementally add features like failover, setsockopt, urgent-data etc. over time? Or, on submissions, do you always want to see a patch series of the full set of features and values according to the SMC design? Kind regards, Ursula